首页 | 本学科首页   官方微博 | 高级检索  
   检索      


Some Results on Error Rates for Quadratic Discrimination with Known Population Parameters
Authors:Dean M Young  Danny W Turner  Virgil R Marco
Institution:1. Baylor University, Waco, TX 76798, USA

Dean M. Young, Department of Information Systems Baylor University Waco, TX 76798 USA

Danny W. Turner, Department of Mathematics Baylor University Waco, TX 76798 USA

Virgil R. Marco, Department of Statistics Oklahoma State University Stillwater, OK 74078 USA;2. Oklahoma State University, Stillwater, OK 74078, USA

Dean M. Young, Department of Information Systems Baylor University Waco, TX 76798 USA

Danny W. Turner, Department of Mathematics Baylor University Waco, TX 76798 USA

Virgil R. Marco, Department of Statistics Oklahoma State University Stillwater, OK 74078 USA

Abstract:Let population Πi, i = 1,2, be characterized by a multivariate normal density function, Ni,∑i), i = 1, 2, respectively. This paper provides conditions under which simple conditional error rates may be computed for the quadratic discriminant function with known population parameters. Also, a simple bound on the overall error rate is derived. Examples are given which demonstrate the proposed methods.
Keywords:Error rates  Singular-value decomposition  Dimension reduction
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号